Purchase this article with an account.
Ben Cipollini, Garrison Cottrell; Asymmetry in lateral connections can account for lateralization of local/global and face processing. Journal of Vision 2013;13(9):1032. doi: 10.1167/13.9.1032.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Lateralization of function touches the things we think make us most human: language, use of our hands, our emotions, and how we perceive the world through our eyes. Current theory postulates that lateralization of visual function is due to an underlying asymmetry in the processing of spatial frequency information (Ivry & Robertson, 1998). However, computational models either fail to describe specific neural mechanisms leading to this asymmetry (Ivry & Robertson, 1998), or do not account for behavioral data showing lateralization of visual function in adults (Howard and Reggia, 2007). Lateralization of visual function appears only under low stimulus strength, suggesting that the postulated asymmetry in spatial frequency processing is in feedback or lateral connections, rather than feed-forward connections. We focus here on lateral connections, which have been well-characterized anatomically, and implement a simple connectionist model to examine how these connections affect spatial frequency processing. We show that small variations in the spatial spread of these sparse, long-range lateral connections can bias spatial frequency processing towards low spatial frequencies. We show that these small variations can arise during development, from an interaction between changes in visual acuity with a hypothesized asymmetry in the timing of visual development (Hellige, 1993). Finally, we show that the asymmetry due to these connection differences can reproduce lateralization of visual function found in humans for a variety of tasks using local/global stimuli (Sergent, 1982), faces (Young & Bion, 1981), and frequency gratings (Kitterle et al, 1992; Christman 1991). Our model accounts for more data than any previous model of lateralization of visual function, postulates no new neuro-developmental mechanisms, and is consistent with a host of independent data related to the development of horizontal connections (Katz & Callaway, 1994), visual asymmetries (Plaut & Behrmann, 2011), and visual processing at multiple scales (Hopf et al, 2006).
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only